Advancements in image sensors and signal processing have led to the successful development of lightweight hyperspectral imaging systems that are critical to the deployment of Photometry and Remote Sensing (PaRS) capabilities in unmanned aerial vehicles (UAVs). In general, hyperspectral data cubes include a few dozens of spectral bands that are extremely useful for remote sensing applications that range from detection of land vegetation to monitoring of atmospheric products derived from the processing of lower level radiance images. Because these data cubes are captured in the challenging environment of UAVs, where resources are limited, source encoding by means of compression is a fundamental mechanism that considerably improves the overall system performance and reliability. In this paper, we focus on the hyperspectral images captured by a state-of-the-art commercial hyperspectral camera by showing the results of applying ultraspectral data compression to the obtained data set. Specifically the compression scheme that we introduce integrates two stages; (1) preprocessing and (2) compression itself. The outcomes of this procedure are linear prediction coefficients and an error signal that, when encoded, results in a compressed version of the original image. Second, preprocessing and compression algorithms are optimized and have their time complexity analyzed to guarantee their successful deployment using low power ARM based embedded processors in the context of UAVs. Lastly, we compare the proposed architecture against other well known schemes and show how the compression scheme presented in this paper outperforms all of them by providing substantial improvement and delivering both lower compression rates and lower distortion.
[1]
J. Makhoul,et al.
Linear prediction: A tutorial review
,
1975,
Proceedings of the IEEE.
[2]
K. Paliwal,et al.
Efficient vector quantization of LPC parameters at 24 bits/frame
,
1990
.
[3]
Wai C. Chu,et al.
Speech Coding Algorithms
,
2003
.
[4]
Shen-En Qian.
Hyperspectral Data Compression Using a Fast
,
2004
.
[5]
Rolando Herrero,et al.
Lossy compression of ultraspectral images: integrating preprocessing and compression stages
,
2014,
Signal Image Video Process..
[6]
Robert M. Gray,et al.
An Algorithm for Vector Quantizer Design
,
1980,
IEEE Trans. Commun..
[7]
William A. Pearlman,et al.
Comparison of 3D set partitioning methods in hyperspectral image compression featuring an improved 3D-SPIHT
,
2003,
Data Compression Conference, 2003. Proceedings. DCC 2003.
[8]
William A. Pearlman,et al.
Three-Dimensional Wavelet-Based Compression of Hyperspectral Images
,
2006,
Hyperspectral Data Compression.
[9]
Philip Ogunbona,et al.
On the computational complexity of the LBG and PNN algorithms
,
1997,
IEEE Trans. Image Process..
[10]
I. Colomina,et al.
Unmanned aerial systems for photogrammetry and remote sensing: A review
,
2014
.
[11]
Michael Mills,et al.
Fractal based image coding scheme using Peano scan
,
1988,
1988., IEEE International Symposium on Circuits and Systems.
[12]
Hiroaki Kuze,et al.
Remote sensing applications with NH hyperspectral portable video camera
,
2012,
Asia-Pacific Environmental Remote Sensing.